These systems consist of three main units - the accelerator, target/blanket and separation units - Fig.1. The accelerator generates high energy (around 1 GeV) charged particles (e.g. protons) which strike a heavy material target. This bombardment leads to the production of a very intense neutron source (a process called spallation). These neutrons enter a subcritical core (often called a blanket) where they can be multiplied. In the core, the transmutation of actinides and fission products takes place. After a time, already transmuted nuclei have to be removed from the fuel in order to avoid their undesirable activation. Long-lived fission products and actinides return from the separation unit to the blanket. Short-lived and stable isotopes as well as fission poisons are removed and processed for storage. It is supposed that introducing ADS can reduce by at least 3 orders of magnitude the time needed for the geological disposal of nuclear wastes.
The idea to use accelerator driven systems is not new. Over the
past few decades, however, the motivation for its development has rapidly
changed. Originally ADS was developed to produce
fissile materials, namely 
Pu from 
U or 
U from
Th in the Material Testing
Accelerators project at the Lawrence Livermoore Radiation Laboratory at
the end of the forties [2].
This project was abandoned in 1952 when high grade uranium ores were
discovered in the United States. In addition, the subsequent idea of
exploiting the spallation process to transmute actinides
and fission products directly turned out to be ineffective.
The proton beam currents required (around 300 mA) were much larger than the
most optimistic theoretical designs for an accelerator could
achieve. Further discoveries in accelerator technology (also a product
of the Strategic Defence Initiative - Star Wars program) have started
a new wave of interest in using ADS. Since the end
of the cold war , another strong impetus to develop ADS has appeared:
the destruction of surplus weapons-grade plutonium.
In the last few years hybrid accelerator-driven systems have been proposed for different purposes. Los Alamos National Laboratory has developed several ideas to use hybrid system with a linear accelerator based on thermal and recently also on fast neutrons for incineration of plutonium and higher actinides, for transmutation of some fission products as well as for gaining energy [3]. Almost four years ago a CERN team lead by Carlo Rubbia proposed a cyclotron based hybrid system to produce nuclear energy based on a thorium fuel cycle. This is a very attractive option reducing the concerns about higher actinides in the spent fuel and offering the possibility of utilising cheap and quite abundant thorium [4].
